News:

Simutrans.com Portal
Our Simutrans site. You can find everything about Simutrans from here.

Real Life Vehicles

Started by Imperior, August 19, 2014, 04:59:26 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Imperior

Hi, it's my first post here, even tought I have been playing this game for like 2 years now, so I'd firstly want to congratulate the developers, it's very well done and gets better every time I download an update.

So, to the issue, I have played many tycoon games during the past years, from Locomotion to Train Tycoon 3(lovely) and OpenTTD and many others that weren't as fun(namely Sid Mier's Railroads). And the thing Simultrans lags behind the most when comparing to any other seems to be it's lack of diversity when it comes to vehicles. I mean, I miss having different sizes of buses in different styles... And also find it awful that a generic Van in 2000 can't reach 60km/h but can fit 25people inside it.

I wonder, if that would be possible in an add-on, or might happen in future updates, or if it just isn't the game's idea... But I'd really find it fun to play around with real life(or at least based on real life) vehicles for trains have been present since 1830 and there are hundreds of models to choose from between 1896-2015(not that one would be crazy to add them all).

A simple suggestion(what have been hurting my gameplay most):

*1830-1850(early trains)
To make the early game more fun, any, there fell like there are few vehicles before ~1940, totally different from Train Tycoon 3 for instance, whose early was fun the most.


*1920-1990(modern bus/van)
For instance, adding one or two new buses, and for instance replacing the generic van with this:

The early Kombe, which was all so common up to the 90s:
http://melhoragora.org/wp-content/uploads/2007/09/k11.jpg

The modern mini-Van, which is an agile small transport very commonly used as "school bus" in my hometown for instance:
http://i01.i.aliimg.com/photo/v193/205521049/Mini_van_JNQ6495E1.jpg

And a longer, larger van like the one presented in the game:
http://sigiritaxi.com/image/van-l.jpg

Hope my ideas are welcome, sorry about the poor images, got them on Google Images.
PS: also, the city growth, it gets really weird to have a 2k city growing faster than a 20k metropolis linear growth is kind of an let down in late game.

IMPORTANT note: beign based on real life vehicles also open doors to use realistic speeds, power, capacity and costs.

Vladki

Perhaps you should try playing different pakset. Those that are country specific have usually real life vehicles. Just try the british, german, czechoslovak or japanese paksets. There are also many addons that you can use with generic paks, e.g. pak hu-pl. And at last if you do not find your favorite vehicle, you can make it yourself.

Sent using recycled electrons.


Ters

All vehicles in pak64, except ships, monorail/maglev, trailers, most wagons and all but a few locomotives (mostly futuristic ones), are real life vehicles. Unfortunately, in my opinion, someone has replaced their names with fictional names (mostly honouring the artist) in the English (and perhaps other) translations. TTD did the same, or so I have heard, compared to standard Transport Tycoon.

Speeds and power are realistic as far as I have seen, probably also weight and to some degree capacity. Cost is fictional, to (try to) fit the fictional game currency that doesn't have inflation.

DrSuperGood

Modern day engine names/looks might be subjected to copyright so I doubt it is fully legal to just add them without first at least asking the copyright holder if it is ok to do so. For this reason I imagine there is some deviation in looks and names from real engines. It can even be the manufactures name which is a problem (eg Alstom) since using that without their permission also has potential legality issues.

As such it is probably best to make engines look "similar" and sound "similar" but not "the same" as real life engines. This is common place in development of economic simulation games where characters like "Gill Bates", and companies like "Mikrosoft" are commonly used.

Obviously does not really apply to very old engines which for all purposes can be considered abandoned or out of copyright. I doubt there is an issue for using the Stephenson's Rocket for example. However there would potentially be quite big issues with using "Alstom Class 390 Pendolino" on the other hand since either Virgin Trains or Alstom might not want it used this way. However I do respect the Pak authors brave enough to ignore such potential legal issues and correctly name their engines after real life engines. It also might not be an issue with open source paksets/programs.

Quote*1830-1850(early trains)
Until Standard gets an industry closure/upgrade/migration model this will never work well. You will end up with a lot of small industries late game that a single truck can supply that are completely worthless compared with 1 single modern industry. This is the biggest problem with pre 1900 play in standard pak sets since either modern industry production amounts are used (you need like 20 parallel roads of carts to even hope to supply a small fraction) or you end up with useless industries late game.

Experimental does this very well since it will start off with low consumption industries and slowly upgrade them. We are still pre 1900 on the server which started in the 1750s. It has other problems however like only ships being able to supply most industries due to production numbers.

The biggest problem with pak64 as it stands is the poor balance. Sadly it is not getting any of the attention it disserves despite being one of the most popular pak sets. Most noticeable are a car transporter from 1960 is more efficient than a truck from post 2000, very high speed trains and maglevs cannot make profits (yes, they physically cannot make profit as they will still lose money 100% full shipping only positively) and that trams are more profitable than trains.

Ters

Quote from: DrSuperGood on August 20, 2014, 02:05:47 PM
Modern day engine names/looks might be subjected to copyright so I doubt it is fully legal to just add them without first at least asking the copyright holder if it is ok to do so. For this reason I imagine there is some deviation in looks and names from real engines. It can even be the manufactures name which is a problem (eg Alstom) since using that without their permission also has potential legality issues.

I have never heard that copyright is an issue, and wouldn't that be a problem for all those sites with collections of train, bus, truck and/or airplane photos as well? Trademarks could be a problem, but the only expaination I've ever come by for as to why all the names were changed between Transport Tycoon and Transport Tycoon Deluxe (or maybe it was for the American release of Transport Tycoon that was sort of in between) was that they feared lawsuits if the game portrayed the vehicles negatively, such as broken down or involved in an accident. That is not an issue for Simutrans. And if the Simutrans community knew of such problems, it would purge all "official" pak sets, not just a few.

prissi

The renaming was done several times on the translator and I got bored to rename them all back. The german translation has mostly their original names.

The other thing is that there are simply no industries for pre 1900. There is the food chain (farms, windmill, market) and that is it. I have no boats in player colour, and rendering tends to produce ugly stuff at 64 scale (to pixely). pak64 lacks fishing and other ship for instance, especially older ones.

The maglevs made between 2,83 and 3,42 cr/tile fully loaded or 750 per 8 car train on a ~100 tile trip. But it may feel too low. Still, a maglev line for a single train will never be profitable (and it is not intended to be too). Maglev are internally balance like airplanes, which are also barely profitable. Anyway, I reduced the cost so a 8 car train running fully loaded twice a month gets its maintance back in 2040.

The problem with the car transporter is that cars are so heavy. So a general purpose truck has its troubles with the trailer without destroying the balance for most other goods. The only way out is to make car transport more expensive and then increase the other prices. However, this will level all transport revenue more or less completely, which takes out the diversity (and hence some of the fun for me) of some good chains.

But I get very rarely feedback apart from those late trains (which I tested a few years back extensively.)

Ters

Quote from: prissi on August 20, 2014, 11:24:34 PM
But I get very rarely feedback apart from those late trains (which I tested a few years back extensively.)

I really wonder how you got the Pantheress and Tigress to make a profit (with the corresponding cars). A few years back was perhaps before speed bonus became based on actual speed, rather than the lowest upper speed limit of the vehicles. Although according to my calculations, even in the latter case, the trains must be quite long compared to other trains in order to break even. Pantheress and Tigress with double decker cars should be quite profitable (at somewhat long, but reasonable lengths), but they wouldn't be high speed anymore.

DrSuperGood

#7
QuoteThe maglevs made between 2,83 and 3,42 cr/tile fully loaded or 750 per 8 car train on a ~100 tile trip. But it may feel too low. Still, a maglev line for a single train will never be profitable (and it is not intended to be too). Maglev are internally balance like airplanes, which are also barely profitable. Anyway, I reduced the cost so a 8 car train running fully loaded twice a month gets its maintance back in 2040.
There are no maglev in pak64 as far as I can tell. All the engines under the "Maglev" section are "Monorail".

The problem is that they cannot make any money end game in Pak64. This is due to the following problems...

1. The fastest engine, "Concept 2" cannot obtain its maximum speed when fully loaded in a passenger configuration. The smallest possible convoy (3 units) can only obtain 584/590 km/h. This is made worse with convoy length with a maximum size (24 unit) only obtaining 540/590. It should be noted that most of that speed loss comes about with the first few units. Maximum speed can still be obtained when not fully loaded but with the run costs that is not something you want to do. The solution is to raise power per unit from 4000 kW to 5000 kW which would suffice. However looking at the units in detail you will realise that the front engine has gearing yet the rest of the coaches do not. If all coaches had the gearing the train would easily reach maximum speed without any issues at all. Is the lack of gearing on all but the front units intentional or just an oversight (makes no sense why front unit should be more powerful than rest)? If it was an oversight this would certainly explain a lot.

2. None of the engines are "Maglev" they are all "Monorail" as far as I can tell (they all have waytype "monorail_track"). In fact I do not think there are any true "Maglev" engines in pak64, they are all Monorail engines. Monorail has an awful speed bonus compared with Maglev. Assuming that the trains can reach their maximum speed (if issue 1 is resolved) the following is the per unit per tile revenue.
If maglev:
Kirayami X-1 : 160 km/h : P 0.11 M 0.13
Concept 1 : 300 km/h : P 0.26 M 0.32
Concept 2 : 590 km/h : P 0.59 M 0.72

If monorail:
Kirayami X-1 : 160 km/h : P 0.06 M 0.07
Concept 1 : 300 km/h : P 0.18 M 0.22
Concept 2 : 590 km/h : P 0.42 M 0.51

Assuming it is monorail (since I cannot see it declaring the engines as maglev anywhere?) and that the engines can reach their intended top speed. A 24 unit Concept 2...
2150 passengers @ 590 km/h for 840.00 money/km (now 672.00 money/km).
Total income from passengers is
2150 * 0.42 = 903

So it did make money! Oh wait it never reached the 590 km/h speed bonus in the first place... Again problem 1 makes a huge difference (and explains everything?).

Per passenger profit...
Old = 0.02930232558139534883720930232558, yeh a tram/plane makes more so this is not very good.
New = 0.10744186046511627906976744186047, not bad. However...

By Plane: RvG A380 = 0.06447058823529411764705882352941 so yes, it makes more profit than a plane!
By Tram: Autocar GT-6 = 0.11043478260869565217391304347826, so why use a maglev when you will make more profit sending them by tram... Tram infrastructure and tunnels is dirt cheap compared with monorails.

3. There is a huge problem with restrictions on construction of the rails for the high speed monorail engine. There are no monorail tunnels and no elevated maglev. You are basically forced into building them above ground at ground level. Maybe raise the elevated maglev maximum speed to 750 as well just to open elevated lines as an option?

4. Why is the way type limited to 750 km/h yet the best engine to 590 km/h? Harmless I admit but still does not look very polished. Until a 750 km/h engine is added lower the rail speed to 590 km/h. Or is this so they can go very fast down hill?

So what is the result of all this?
1. Fix the speed issue as a result of gearing, reaching maximum speed helps a lot! Before they could not make profit at their actual speeds.
2. Either lower running cost even further to like 25 per car or raise running cost of trams. Considering you are paying 200 per tile you would expect them to make more money than paying 80 per tile trams.
3. Increase the maximum speed of elevated maglev track to that of ground level maglev to open up new construction options. Some kind of maglev tunnel would also be nice but at least elevated ways would allow above ground freedom of construction.
4. Decrease the maximum speed to that of the fastest engine until a faster engine is added so people do not expect a faster engine to eventually be available.

In any case the run cost reduction is a huge improvement. Next release I will actually try using them as the difference between trams might just be worth it for the speed.

I will discuss the high speed rail trains tomorrow.

prissi

Histrically there was only a rail called monoral, which became later (when Timothy drew hes vehicles) the maglev. In order to load older games this has to stay monorail waytype. So there is indeed not "true" maglev.

In my tests the concept 2 reached near maximum speed in units of 8. But that can depend on the fast forward settings.

You can use cars.xls from the SVN to get all the costs.

The tram is much slower and much less capacity. Trams are for starting. If you need to move 50000 or more per month, then think maglev. If there are less than 1000: trams ... (Same in pak64 Shinkansen.)

The 750 was for future cars, which were never made. I would like to seem more variety in the depot ...

I will check the gearing issue. Also the elevated lines in red should have the same 750 km/h speed limit. There won't be tunnels as for all existing lines crossing speed above 500 km/h creates sonic booms in tunnels. The japanese test for this just up to 500 km/h.

Ok, Pathress and Tigress also fixxed (mostly be removing the last increase of speedbonus).

DrSuperGood

#9
QuoteIn my tests the concept 2 reached near maximum speed in units of 8. But that can depend on the fast forward settings.
I opened up the RC version 120, started a new game in the year 2222 (to make sure everything is unlocked) and then built a monorail depot on some maglev way type. I then began assembling a "Concept 2" using the front unit, 22 passenger modules and the end unit to max out at 24 units. The quoted max speed was not the convoy maximum.

Edit I think you need to give coaches gearing as well. They cannot sufficiently power themselves at maximum speed.

QuoteThe tram is much slower and much less capacity. Trams are for starting. If you need to move 50000 or more per month, then think maglev. If there are less than 1000: trams ... (Same in pak64 Shinkansen.)
Except trams I can stack 2, 4 or even 8 tiles deep underground in road tunnels. I cannot even build high speed elevated maglev. I did this on your servers a while back when my underground loops could not pick up enough people to my central stations.

QuoteI will check the gearing issue.
Edit: Concept 1 and 2 both suffered from it I think. Also unless you want to enforce a maximum convoy length at maximum speed you may wish to give passenger and mail coaches some gearing so they are at least fully self-propelled at full speeds.

QuoteAlso the elevated lines in red should have the same 750 km/h speed limit.
I am glad to hear this. It will make maglev/monorails so much more usable and could even justify the non-existence of tunnels for this type (part of the challenge of using them). Having some building options is exactly what they need.

QuoteThere won't be tunnels as for all existing lines crossing speed above 500 km/h creates sonic booms in tunnels. The japanese test for this just up to 500 km/h.
Its the year >2040... Low pressure/vacuum tunnels anyone? Not important anyway as the other fixes are more than enough to bring them into gameplay.

QuoteOk, Pathress and Tigress also fixxed (mostly be removing the last increase of speedbonus).
Is this wise?! You will have just greatly increased the end game profit from freight as well as everything uses that speed bonus. Coal will give like 0.02 more per km. On a lesser note shipping cars and food by train should now be profitable again (exploding with joy)! Maybe this is for the better but you will probably need to nerf some older trains a tad.

Also will it make more profit than the N-ICE middle speed diesel engine? I will investigate.

Ters

Quote from: DrSuperGood on August 21, 2014, 11:51:45 PM
Its the year >2040... Low pressure/vacuum tunnels anyone?

Then the entire network has to be that way. That would be a completely separate waytype.

Quote from: DrSuperGood on August 21, 2014, 11:51:45 PM
Is this wise?! You will have just greatly increased the end game profit from freight as well as everything uses that speed bonus. Coal will give like 0.02 more per km. On a lesser note shipping cars and food by train should now be profitable again (exploding with joy)! Maybe this is for the better but you will probably need to nerf some older trains a tad.

Wouldn't it just put freight prices back where they were in 1999? Freight trains didn't get any faster in conjunction with the last speed bonus speed increase. The powerful high speed locomotives might be able to pull more goods at maximum freight speed, though.

DrSuperGood

Could we please revise rail tunnels? They have been bugging players on the servers a lot due to how useless they are.

1. 120 km/h rail tunnel obsoletes at 01/2000. Who would have thought that one might actually want to use a tunnel for moving goods or slow passengers which do not need high speeds. Currently we are forced to use tram ways in road tunnels as there is no way it is economical to build the high speed tunnels. Please remove the obsolete date from 120 km/h rail tunnels as that speed is still useful throughout the game. Alternatively a replacement low speed tunnel with better properties could be added but this may require more art assets.

2. 120 km/h rail tunnels cost far too much to maintain. For 140 money/month per tile I can build a road tunnel (100 money/month for 80 km/h) with tram way (40 money/month for 90 km/h) in it allowing both cars and trains simultaneously. On the other hand I could make a 120 km/h rail tunnel costing 200 money/month per tile that can only move trains. The 60 money/month more is not worth the loss of road traffic and gain of 30 km/h rail speed (for a purpose speed does not even mater much). As such I would strongly recommend lowering the maintenance cost of 120 km/h rail tunnel to 100 money/month (2500 in game data), giving faster speeds than same value of road which is tradition in Simutrans.

3. High speed rail tunnels are not very high speed. At 300 km/h they cannot support the 450 km/h that above ground rails can. This may cause them problems with high speed passenger traffic forcing purely above ground routes (which becomes too competitive against maglev which should dominate that area). Please make them 450 km/h like Steel Sleeper Track so that they are completely compatible with your high speed network.

4. As with 2 and especially with 3, High Speed Rail Tunnels cost far too much to maintain. At 600 money/month they have the highest maintenance cost of any way type in pak64 standard. More expensive than even airport runways something is seriously wrong with their maintenance. To put it in perspective you can get 6 tiles of high speed bridge for the same cost as 1 tile of high speed tunnel and that does not have speed limit issues. In line with cost reduction suggestion 2 I would recommend lowering them to 320 money/month (8000 game data maintenance). This value was chosen to be 5 times the above ground rail maintenance cost to reflect the new 120 km/h tunnel maintenance (which is also 5 times, which also reflects high speed road tunnels).

5. As mentioned in 4, something is seriously wrong with the high speed rail bridges maintenance cost. At 100 money/month they cost the same maintenance as a lower speed bridge and with no length restrictions this equivalently obsoletes the "Steel girder rail bridge". This can only be the result of an error. As I cannot compute a maintenance price for it, I must suggest simply raising it to 150 money/month (3750 game data).

kierongreen

paksets should be strongly balanced against building tunnels so to me the current values for pak64 seem ok (in fact, if the rail is expensive compared to road I'd say raise the price of the road rather than lowering that of railways). Tunnels should result in losses when trains use them unless you have very, very high utilisation. When deciding on a route you then have to only use tunnels when you really need to.

DrSuperGood

Quotepaksets should be strongly balanced against building tunnels so to me the current values for pak64 seem ok (in fact, if the rail is expensive compared to road I'd say raise the price of the road rather than lowering that of railways). Tunnels should result in losses when trains use them unless you have very, very high utilisation. When deciding on a route you then have to only use tunnels when you really need to.
No they should only be strongly balanced against them if it is the pak set author's desire that they should not be used (like with maglev in pak64 where there are no tunnels at all). In network play tunnels are a god send since play would otherwise be unbearable as you desperately try to navigate a tangled mess of other companies ways which act as huge construction obstacles. Elevated ways do not help since their stops block them for some silly reason (bad build logic? they can make stops under your elevated ways which is also annoying).

As pak64 is one of the most commonly played network pak sets I would recommend keeping tunnels as a viable alternative for most kinds of traffic (like it is currently with road tunnels). I would not mind a sharp cost rise however (10 times construction cost) since as it is you end up with billions in your coffers and this would migrate tunnels to a late game way type (when money starts to flow in).

Ters

Quote from: DrSuperGood on August 22, 2014, 02:56:36 PM
3. High speed rail tunnels are not very high speed. At 300 km/h they cannot support the 450 km/h that above ground rails can. This may cause them problems with high speed passenger traffic forcing purely above ground routes (which becomes too competitive against maglev which should dominate that area). Please make them 450 km/h like Steel Sleeper Track so that they are completely compatible with your high speed network.

Tunnels are not compatible with high speed networks in real life. That's just how it is, and it's part of the challenge. With a speed limit of 300 km/h, the tunnel in pak64 is at the edge of what's realistic according to all information I can find.

DrSuperGood

QuoteTunnels are not compatible with high speed networks in real life. That's just how it is, and it's part of the challenge. With a speed limit of 300 km/h, the tunnel in pak64 is at the edge of what's realistic according to all information I can find.
Someone above said a speed of 500 km/h in tunnels was the max for maglev tunnels and obtained in Japan.

However I thought the Tigress type engines (the high speed late game engines from the future) were all made up anyway. Just like the 2 tile (km?) wide double tracks, the 12 km long engines, more passengers being in transit in a stop than on the map and basically a lot of things. I personally would not mind sacrificing a tad of realism with regards to tunnels just to improve the gameplay in crowded situations.

In real life the main reason tunnels are not used more often is the high construction cost, geographical restrictions, lower safety (accidents in a tunnel are a lot more dangerous than out of one) and that they take forever to build. Once they are built I would not image tunnels having that high a maintenance cost (5-6 times seems reasonable). My biggest complaint with them at the moment is they are too easy to build. In real life a city underground system takes the order of decades to construct and costs billions. In simutrans you can build the same system instantly (game paused) and it costs only a million or two.

As such a 10-50 times increase to build costs with tunnels might be a good thing rather than making them unviable as a method of transport.

prissi

Firest, thank you for you observation. These are most helpful to get the pak right again.

pak64 was made against using tunnels too much, and on most server games also altering the landscape is very expensive (at least on the one I ran). But you are right, the relative costs must be correct.

The original balance was 2.5x for bridges and 5x maintenance for tunnels. I restored these values.

Ters

Quote from: DrSuperGood on August 22, 2014, 05:53:42 PM
Someone above said a speed of 500 km/h in tunnels was the max for maglev tunnels and obtained in Japan.

All I could find was that the Japanese shelved a project to upgrade to 360 km/h, mostly because of problems with sonic booms in tunnels. They have apparently begun looking into it again, but it is still just something of the future at best.

Quote from: DrSuperGood on August 22, 2014, 05:53:42 PM
However I thought the Tigress type engines (the high speed late game engines from the future) were all made up anyway. Just like the 2 tile (km?) wide double tracks, the 12 km long engines, more passengers being in transit in a stop than on the map and basically a lot of things. I personally would not mind sacrificing a tad of realism with regards to tunnels just to improve the gameplay in crowded situations.

In real life the main reason tunnels are not used more often is the high construction cost, geographical restrictions, lower safety (accidents in a tunnel are a lot more dangerous than out of one) and that they take forever to build. Once they are built I would not image tunnels having that high a maintenance cost (5-6 times seems reasonable). My biggest complaint with them at the moment is they are too easy to build. In real life a city underground system takes the order of decades to construct and costs billions. In simutrans you can build the same system instantly (game paused) and it costs only a million or two.

As such a 10-50 times increase to build costs with tunnels might be a good thing rather than making them unviable as a method of transport.

I have never found tunnels to be unviable, except perhaps for the cost (I mostly need tunnels when I can't yet afford them).

And keep in mind that Simutrans has a symbolic representation of the world, as do most, if not all, transportation games. Roads and vehicles are bigger than life simply because they wouldn't be visible otherwise. Trains also move up something like 30 degree slopes, which in my opinion looks even sillier in Railroad Tycoon 3 when the terrain is less symbolic. These are not the things Simutrans is about simulating. Building a transportation network is, so having tunnels as a bottleneck makes perfect sense to me. They are an evil that might be a lesser evil at times.

DrSuperGood

Quoteand on most server games also altering the landscape is very expensive (at least on the one I ran).
You have no idea... On some of your server games I was so rich I literally made artificial islands in the middle of the ocean just because I could (well another player suggested it and I thought I would give it a try, I was still very new).

10k per tile? No problems when I am earning 300M per year. Thus why a sharp cost increase to tunnels might be appropriate since then you can at least push them back to when people are desperate or just trying to maximize throughput on an overcrowded map.

QuoteThey are an evil that might be a lesser evil at times.
I do not understand why people keep calling tunnels evil. I do agree that their construction for benefit is a bit too easy but I personally love them. When I cannot build flat, I build up. When building up is not enough (a city, bad terrain or just other players) I build under. I am glad now that my tram systems can use 120 km/h tunnels instead of 90 km/h tramways in 80 km/h tunnels. With the buff to speed bonus late game I might not even have to use trams anymore (120 km/h trains!). This is what I have always wanted to see and I cannot wait.

QuoteThe original balance was 2.5x for bridges and 5x maintenance for tunnels. I restored these values.
I like the changes to 120 km/h tunnels (cannot wait to see them on a server) however you appear to have missed out 300 km/h high speed tunnels.

Ters

Quote from: DrSuperGood on August 22, 2014, 09:48:07 PM
I do not understand why people keep calling tunnels evil.

You called them evil, although without using the word, because they slowed down trains you wanted to go as fast as without tunnels. I don't see them as evil, they are a trade-off of top speed versus several other things. Having to go over the mountain might be even slower. Cutting through a city would be too expensive and disruptive to the city.

DrSuperGood

QuoteYou called them evil, although without using the word, because they slowed down trains you wanted to go as fast as without tunnels. I don't see them as evil, they are a trade-off of top speed versus several other things. Having to go over the mountain might be even slower. Cutting through a city would be too expensive and disruptive to the city.
I do not care about physical speed, only what I am paid for. If the way type limits speed then this affects speed bonus resulting in you getting paid less. Only in experimental is actual average speed important (so a slow tunnel may be faster than climbing a mountain). I want the tunnel to be capable of maximum speed so that you get paid for it being maximum speed. Due to passenger high speed bonus and the huge running cost of pak64 vehicles it is important to reach maximum speed as otherwise your profits get eaten heavily. (even shipping 1 in every 10 tiles not positively will waste most of the profit).

In Pak128 we have 400 km/h tunnels, why in pak64 are they limited to 300 km/h? Yes they are affordable in Pak128, probably more so than in Pak64.

I am not asking to make tunnels rigged. I am just asking that high speed rail tunnels are given the same speed cap as high speed steel sleeper rails (450 km/h instead of 300 km/h) and have their maintenance cost lowered in line to all other tunnels (5 times the cost of surface level way of the same speed). Alternatively an extra new high speed tunnel "very high speed" could be made and introduced towards 2000 but this would need new art assets so the easiest fix is to just promote existing high speed tunnels to be useful.

Ters

Quote from: DrSuperGood on August 23, 2014, 04:22:37 PM
In Pak128 we have 400 km/h tunnels, why in pak64 are they limited to 300 km/h? Yes they are affordable in Pak128, probably more so than in Pak64.

I wonder why pak128 has 400 km/h tunnels. Why pak64 stops at 300 km/h seems well founded in reality.

Quote from: DrSuperGood on August 23, 2014, 04:22:37 PM
I am not asking to make tunnels rigged. I am just asking that high speed rail tunnels are given the same speed cap as high speed steel sleeper rails (450 km/h instead of 300 km/h) and have their maintenance cost lowered in line to all other tunnels (5 times the cost of surface level way of the same speed). Alternatively an extra new high speed tunnel "very high speed" could be made and introduced towards 2000 but this would need new art assets so the easiest fix is to just promote existing high speed tunnels to be useful.

And I'm just asking for tunnels to remain a slow alternative to making a deep cutting. Cheaper to build, but slower and more costly to maintain. Although cuttings should realistically have some increased maintainance costs, this might be difficult to implement.

If anything is putting an unrealistic dent in my speed bonus, it's the impact of having to make a slight deviation from a straight line. There is nothing inbetween a sharp 45 degree bend and no bend. That is however a consequence of the game being tile based that is very difficult to avoid.

Philip

Quote from: Ters on August 23, 2014, 06:03:39 PM
If anything is putting an unrealistic dent in my speed bonus, it's the impact of having to make a slight deviation from a straight line. There is nothing inbetween a sharp 45 degree bend and no bend. That is however a consequence of the game being tile based that is very difficult to avoid.

Experimental keeps track of the last few tiles' travel direction so a 45 degree bend surrounded by straight tiles should have much less of an impact (or none at all?) than two 45 degree bends following each other. The code is in simvehikel.cc:calc_speed_limit; I'm not sure it's configured properly so a single tile deviation indeed has no impact at all, but the difficult work is done.

Ters

Quote from: Philip on August 23, 2014, 06:18:01 PM
Experimental keeps track of the last few tiles' travel direction so a 45 degree bend surrounded by straight tiles should have much less of an impact (or none at all?) than two 45 degree bends following each other.

Standard also treats two 45 degree bends harsher than one 45 degree bend, but in my example, there are two 45 degree bends following each other, one turning either way. In this particular case, the two 45 degree bends should have less impact than one 45 degree bend, but in other cases, which the game can't really tell apart, two 45 degree bends should have more impact than one 45 degree bend.

Philip

Quote from: Ters on August 23, 2014, 08:46:42 PM
Standard also treats two 45 degree bends harsher than one 45 degree bend, but in my example, there are two 45 degree bends following each other, one turning either way. In this particular case, the two 45 degree bends should have less impact than one 45 degree bend, but in other cases, which the game can't really tell apart, two 45 degree bends should have more impact than one 45 degree bend.

I don't understand why you think the game cannot tell those two situations apart, can you elaborate? The way the current code works is by comparing each of the last n travel directions to the current direction, then remembering the maximum difference—45 in your case, 90 if there are two 45-degree turns in the same direction on that stretch.

I think turns with a maximum difference of 45 should be free: you can turn around a TGV at full speed, it just requires a really large turning circle. More realistically, a 45-degree-and-back section is just how we simulate ways going in intermediate directions. The attached one-liner patch (which probably applies to Experimental at an offset because I've made other changes to the file) does that, and also fixes the minor bug that if corner debugging was enabled, the top speed of vehicles on straight ways would also be reduced. It appears to work as intended, but I've only tested it very briefly.

Ters

Quote from: Philip on August 23, 2014, 10:03:10 PM
I don't understand why you think the game cannot tell those two situations apart, can you elaborate?

It's all because of the minimum granularity of the Simutrans world. In the real world, a high speed line having to deviate from a straight line between two stations can do so with curves stretching over many kilometers if there are no other obstacles. In Simutrans however, any deviation must be at least one tile to the side, and the shift will happen over a distance of one tile. So what is supposed to be a minor deviation becomes just the same as a track navigating through hilly terrain. I try to illustrate with these to images, one of a high speed track that has to shift slightly from perfectly straight, the other is a track having to make two sharp turns to curve along the coast at the bottom of mountain (unfortunately hard to see, but I didn't have much time to prepare the images). However the track is exactly the same (ignore that neither actually uses the actual high speed track from pak64), just the terrain is different.

DrSuperGood

In standard as far as I can tell each 45 degree bend adds drag to the coaches being pulled across those tiles. Having 2 such bends in a row will add twice the drag from coaches independent of direction. A 90 degree bend adds a lot more drag than 2 45 degree bends. By limiting your convoys to only 1 bend at any given time your excess power can compensate to some degree for the bend (minimal speed impact so faster turning).

What does this mean as far as game play? When going around a corner your train is given a new slower maximum speed (which does not alter speed bonus) as a result of the drag making the coaches appear more heavy. This is how a very fast train can still accelerate around a corner when leaving a station and even reach >200 km/h however it will not max out as the extra drag prevents it from obtaining its maximum speed physically (as if it weighed more).

QuoteI wonder why pak128 has 400 km/h tunnels. Why pak64 stops at 300 km/h seems well founded in reality.
For the reason I mentioned before, gameplay.

The fastest trains in pak128 (no maglev in it so only high speed rail) run at 400 km/h and that is the tigress introduced around 2050 (oh fifty, we never did get to use them on the second pak128 server :( ). To allow you to use them they give you options of surface rail (cheapest), bridge and elevated (more expensive), and tunnel (most expensive) all capable of letting the engine reach its 400 km/h top speed. It is electric so all of those way types need the expensive 400 km/h rated high-speed overhead lines.

In pak64 no one has really used high speed trains recently for the reason everyone has complained about which is now fixed (they could not break even) and so the lack of a max speed capable tunnel was not an issue as there were no viable trains that fast anyway. Now with them being fixed we really need viable high speed tunnels. For this reason I would strongly recommend raising them to 450 km/h to match steel sleeper track and then lowering their maintenance to comply to the 5 times surface way standard the pakset uses. Even if such use of tunnels is not realistic, it is justified simply to make the pakset more multiplayer friendly since when playing with other players it is often not possible to push their lines out the way for your high speed surface line.

Philip

Quote from: Ters on August 23, 2014, 10:37:17 PM
It's all because of the minimum granularity of the Simutrans world. In the real world, a high speed line having to deviate from a straight line between two stations can do so with curves stretching over many kilometers if there are no other obstacles. In Simutrans however, any deviation must be at least one tile to the side, and the shift will happen over a distance of one tile. So what is supposed to be a minor deviation becomes just the same as a track navigating through hilly terrain. I try to illustrate with these to images, one of a high speed track that has to shift slightly from perfectly straight, the other is a track having to make two sharp turns to curve along the coast at the bottom of mountain (unfortunately hard to see, but I didn't have much time to prepare the images). However the track is exactly the same (ignore that neither actually uses the actual high speed track from pak64), just the terrain is different.

Thank you! I don't feel it's all that important to penalize the hilly track—if you can navigate through terrain using only 45-degree turns, without having to do a left-right-right-left sequence within (I believe) 10 tiles, it's not actually that hilly. It would be possible, if really necessary, to check that the "overshoot" tiles are at the right height before allowing free 45-degree turns, but that would, I think, cause more trouble than it's worth. Note that if there's a sequence of 45 degree turns going left-right-right-left within a few tiles, that will count as a 90 degree turn, which I think should cover most actual cases of hilly terrain. I might be wrong, though.

Ters

The thing might be that in my hilly example, it shouldn't be two 45 degree turns, but two 90 degree turns (or even two 135 degree turns). It's just that those 90 degree turns are so close that there isn't room for a full straight tile between them.

DrSuperGood

In the case of hilly track I would recommend either flattening the hill or digging a channel through it (low maintenance, huge construction, often ugly) or making a tunnel. I also recommend for high speed lines not having more than 1 45 degree bend every convoy length to make sure that their physical speed is affected as little as possible. In pak128 you may wish to fork out on extra engine power for lines involving regular sharp turns so that speed is affected even less however in pak64 this is not very viable due to how little profit you make compared with run cost.

Ters

Some players prefer a scenic route over a efficient route.